We present a systematic evaluation of large language models on quantum mechanics problem-solving. Our study evaluates 15 models from five providers (OpenAI, Anthropic, Google, Alibaba, DeepSeek) spanning three capability tiers on 20 tasks covering derivations, creative problems, non-standard concepts, and numerical computation, comprising 900 baseline and 75 tool-augmented assessments. Results reveal clear tier stratification: flagship models achieve 81\% average accuracy, outperforming mid-tier (77\%) and fast models (67\%) by 4pp and 14pp respectively. Task difficulty patterns emerge distinctly: derivations show highest performance (92\% average, 100\% for flagship models), while numerical computation remains most challenging (42\%). Tool augmentation on numerical tasks yields task-dependent effects: modest overall improvement (+4.4pp) at 3x token cost masks dramatic heterogeneity ranging from +29pp gains to -16pp degradation. Reproducibility analysis across three runs quantifies 6.3pp average variance, with flagship models demonstrating exceptional stability (GPT-5 achieves zero variance) while specialized models require multi-run evaluation. This work contributes: (i) a benchmark for quantum mechanics with automatic verification, (ii) systematic evaluation quantifying tier-based performance hierarchies, (iii) empirical analysis of tool augmentation trade-offs, and (iv) reproducibility characterization. All tasks, verifiers, and results are publicly released.
We present a systematic evaluation of large language models on quantum mechanics problem-solving. Our study evaluates 15 models from five providers (OpenAI, Anthropic, Google, Alibaba, DeepSeek) spanning three capability tiers on 20 tasks covering derivations, creative problems, non-standard concepts, and numerical computation, comprising 900 baseline and 75 tool-augmented assessments. Results reveal clear tier stratification: flagship models achieve 81% average accuracy, outperforming mid-tier (77%) and fast models (67%) by 4pp and 14pp respectively. Task difficulty patterns emerge distinctly: derivations show highest performance (92% average, 100% for flagship models), while numerical computation remains most challenging (42%). Tool augmentation on numerical tasks yields task-dependent effects: modest overall improvement (+4.4pp) at 3x token cost masks dramatic heterogeneity ranging from +29pp gains to -16pp degradation. Reproducibility analysis across three runs quantifies 6.3pp average variance, with flagship models demonstrating exceptional stability (GPT-5 achieves zero variance) while specialized models require multi-run evaluation. This work contributes: (i) a benchmark for quantum mechanics with automatic verification, (ii) systematic evaluation quantifying tier-based performance hierarchies, (iii) empirical analysis of tool augmentation trade-offs, and (iv) reproducibility characterization. All tasks, verifiers, and results are publicly released.
We present a systematic evaluation of large language models on quantum mechanics problem-solving. Our study evaluates 15 models from five providers (OpenAI, Anthropic, Google, Alibaba, DeepSeek) spanning three capability tiers on 20 tasks covering derivations, creative problems, non-standard concepts, and numerical computation, comprising 900 baseline and 75 tool-augmented assessments. Results reveal clear tier stratification: flagship models achieve 81% average accuracy, outperforming mid-tier (77%) and fast models (67%) by 4pp and 14pp respectively. Task difficulty patterns emerge distinctly: derivations show highest performance (92% average, 100% for flagship models), while numerical computation remains most challenging (42%). Tool augmentation on numerical tasks yields task-dependent effects: modest overall improvement (+4.4pp) at 3x token cost masks dramatic heterogeneity ranging from +29pp gains to -16pp degradation. Reproducibility analysis across three runs quantifies 6.3pp average variance, with flagshipmodels demonstrating exceptional stability (GPT-5 achieves zero variance) while specialized models require multi-run evaluation. This work contributes: (i) a benchmark for quantum mechanics with automatic verification, (ii) systematic evaluation quantifying tier-based performance hierarchies, (iii) empirical analysis of tool augmentation trade-offs, and (iv) reproducibility characterization. All tasks, verifiers, and results are publicly released.
We adapt Microsoft's QuantumKatas -- a well-established quantum computing curriculum -- from Q# to Qiskit, the most widely-adopted quantum computing framework, and package it with an evaluation framework for systematic LLM assessment. The resulting benchmark comprises 350 tasks across 26 categories, spanning fundamental gates through advanced algorithms (Grover's, Simon's, Deutsch-Jozsa), error correction, key distribution, and quantum games. Each task includes a natural language prompt, canonical solution, and deterministic test verification via classical circuit simulation. By building on the QuantumKatas' proven pedagogical design rather than creating tasks from scratch, we inherit a principled difficulty progression and comprehensive concept coverage while contributing the framework adaptation, evaluation infrastructure, and empirical analysis. We evaluate 16 LLMs across 7 prompting configurations -- a total of 39,200 model runs -- to demonstrate the benchmark's utility. Three key findings emerge: (1) the benchmark effectively differentiates model capabilities, with best-configuration pass rates ranging from 32.3% to 83.1% and a 26.1 pp average gap between frontier and open-source models; (2) models perform well at implementing known algorithms (SimonsAlgorithm 82.1%, BasicGates 81.6%) but struggle with problem encoding (SolveSATWithGrover 34.4%, DistinguishUnitaries 40.0%); and (3) chain-of-thought prompting shows a modestly bimodal effect -- it is the best strategy for three models (two of them explicitly reasoning-tuned per vendor documentation) but degrades performance for the rest, leaving it mid-pack in aggregate (56.3% mean) behind few-shot-5 (57.8%). We release the benchmark, evaluation framework, and baseline results to support research on LLM capabilities in quantum computing.
Abstract The application scenarios of large language models in other fields, such as physics, are still being explored. Typically, a majority of these types include formal mathematical expressions, precise numerical calculations, and other specialised symbols at the same time. Large language models have higher requirements for this problem, and they are mainly built on the basis of natural language model training. Based on current evaluations, it is not entirely clear what these models' reliability and capability limits are at this stage in this field. To address the structural problem of "lack of formal computational verifiability" and the defect of "multiple-choice assessment unable to reflect true reasoning ability" in quantum science problems, a multi-angle extended evaluation system is proposed in this paper, which rigorously reconstructs the QuantumBench test standard. We introduce a new kind of Reasoning mechanism based on the combination of selective symbolic augmentation here. This method limits the use of external symbolics to high-yield sub-distributions while retaining evaluation consistency, making it suitable for a systematic study of the boundary conditions for tool-assisted applications in quantum tasks. At the same time, we build an open-ended Quantum Reasoning Task Suite and propose a two-level evaluation framework comprising regularised automatic judgement and LLM-as-a-Judge that breaks down this relationship. Additionally, we Design a graduate-level multi-stage quantum reasoning benchmark to explore the maximum capacity of existing models for high-order theoretical deduction. Experiments show that a single multiple-choice format struggle to capture the full reliability of the model for Quantum Reasoning. In contrast, the proposed extended evaluation framework systematically reveals systemic differences across varying reasoning paths, tool usage, and task difficulty levels, providing more diagnostic applications for evaluating and deploying large language models in quantum domains.
We demonstrate that large language model (LLM) agents can autonomously perform tensor network simulations of quantum many-body systems, achieving approximately 90% success rate across representative benchmark tasks. Tensor network methods are powerful tools for quantum simulation, but their effective use requires expertise typically acquired through years of graduate training. By combining in-context learning with curated documentation and multi-agent decomposition, we create autonomous AI agents that can be trained in specialized computational domains within minutes. We benchmark three configurations (baseline, single-agent with in-context learning, and multi-agent with in-context learning) on problems spanning quantum phase transitions, open quantum system dynamics, and photochemical reactions. Systematic evaluation using DeepSeek-V3.2, Gemini 2.5 Pro, and Claude Opus 4.5 demonstrates that both in-context learning and multi-agent architecture are essential. Analysis of failure modes reveals characteristic patterns across models, with the multi-agent configuration substantially reducing implementation errors and hallucinations compared to simpler architectures.
Large language models (LLMs) can generate quantum programs, but reliable evaluation remains difficult because benchmarks differ in task composition, scoring, and failure handling. We evaluate 15 hosted frontier, open-weight, quantum-specialized, and retrieval-augmented models on five quantum code-generation suites (530 tasks), plus 10 models on our new QuantumBenchEval (QBE) benchmark (96 tasks across six topics). We report bootstrap confidence intervals, multiplicity-adjusted paired comparisons, and audits of output validity and evaluator behavior. QiskitHumanEval-Standard approaches saturation (98.7% pass@1), while its Hard variant retains headroom (63.8%). None of the five top-two comparisons across primary suites remains significant after multiple-comparison correction. Evaluator design substantially changes scores: removing checks for unspecified quantities raises GPT-6 Astra's QBE pass@1 from 0% to 94.1% (T1) and from 1.2% to 92.9% (T6), without changing model outputs. We identify 17 QuanBench-117 tasks with shared failures linked to benchmark-owned imports. Controlled probes also reveal unexecuted unit tests, leakage of canonical functions into candidate execution, and reference imports that block otherwise valid candidates. These defects can produce both false failures and false passes, so historical QuanBench results require revalidation under an isolated evaluator. Overall, leaderboard scores alone do not adequately characterize progress in quantum code generation. We release our harness, per-sample results, and audit artifacts and recommend validating evaluators, isolating candidate code, and reporting uncertainty and output validity.
We introduce QC-Stark, a benchmark for evaluating large language models (LLMs) on 11 quantum computing (QC) tasks, spanning circuit construction, debugging, compilation, error correction, and simulation. Across 2,750 evaluations (10 models × 11 tasks x 5 difficulty levels x 5 seeds), we find that overall rankings mask substantial per-task variation. The Spearman correlation between overall and per-task rankings is statistically insignificant for 4 out of the 11 tasks included in this benchmark. A 2-parameter Item Response Theory (IRT) model validates measurement quality, and prompt sensitivity analysis confirms ranking robustness across prompt conditions. All tasks are auto-verifiable via execution, thus not requiring any manual evaluation. We make the code and data publicly available on Huggingface.